厦门大学学报（自然科学版）2013，Vol.52Issue(4)：455-458,4.DOI:10.6043/j.issn.0438-0479.2013.04.005
潜在狄利克雷分配模型在网络日志的应用
Web Log Data Based on Potential Lejeune Dirichlet Allocation Model
许两有 1许珠香2
作者信息
- 1. 厦门大学信息科学与技术学院,福建厦门361005
- 2. 厦门网宿科技股份有限公司,福建厦门361008
- 折叠
摘要
Abstract
In recent years,the data mining technology based on the Web log is becoming a hot issue in the theoretical research and commercial applications,and the classification of Web users is one of the most important research in the field of data mining.Classification of Web users can help us find the similarities between the behavior of users,so as to set up the corresponding service project for specific user group.In this paper,according to the history information of the URL which users had accessed,we proposed a user classification method based on weighted potential Lejeune Dirichlet allocation (LDA) model,and the user is divided into different topic groups.Experiments show that this method can achieve very good classification results.关键词
Web日志/潜在狄利克雷分配(LDA)模型/URL/分类Key words
Web log/potential Lejeune Dirichlet allocation(LDA)/URL/clustering分类
信息技术与安全科学引用本文复制引用
许两有,许珠香..潜在狄利克雷分配模型在网络日志的应用[J].厦门大学学报（自然科学版）,2013,52(4):455-458,4.
